Goals for today

  • Understand goals of various regression models
  • Learn ho to select variables for explanative modeling

Previously on Applied Regression in R

What is Regression good for?

If you wish to make an apple pie from scratch, you must first invent the universe.

  • Carl Sagan

Backpedalling Mars?

Mars, what are you doing?

Ptolemaic model

Universe according to Ptolemy

Copernicus

Universe accordin to Copernicus

To Explain or To Predict?

To Explain or To Predict?

Predictive models

  • Predicting the future (or the past!)
  • Interpretation less important.
  • Goal: estimate (unseen) observations as best as possible.

Explanative (causal) models

  • Explaining workings of the universe.
  • Predictive power less important.
  • Goal: Estimate model parameters as best as possible.

To Explain or To Predict?

  • Other differences include: choosing variable, evaluating model fit, choosing sample sizes and more…

  • For more details, see: Shmueli, G. (2010). To Explain or To Predict? (SSRN Scholarly Paper ID 1351252). Social Science Research Network. https://doi.org/10.2139/ssrn.1351252

Other uses of regression models

Descriptive Models

  • Basicaly just a math summarization
  • Goal: Summarize structure of the data

Inferential models

  • Sample to population inference
  • Goal: Describe population as best as possible

What can regression be used for?

  • Predictive models: Which people are going to vote?
  • Explanative models: What is the effect of age on voter turnout?
  • Descriptive models: What is the relationship between age and voter turnout?
  • Inferential models: How many people are going to vote?


- We are going to be mainly interested in explanative models.